ClawHub
Back to skill

Security audit

TGA Analyze

Security checks for vulnerabilities and agentic risk

Overview

This skill appears purpose-built for TGA report downloads, but it requires users to place TGA credentials and browser cookies in local files and stores a reusable token in plaintext.

Review before installing. Use only with a TGA account you are authorized to access, keep the skill directory private, do not commit .env or .tga-token, delete or rotate stored tokens after use, and treat downloaded reports as sensitive business data.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill clearly instructs the agent to perform network access and shell execution, yet it declares no permissions. This undermines the permission model and can cause users or orchestrators to approve a skill without understanding that it will log in to a remote service, execute local commands, and manipulate files.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The documented behavior overstates user-visible analysis functionality while omitting important security-relevant behavior such as persistent token storage and direct credential use from local files. This kind of description-behavior mismatch reduces informed consent and can cause an agent to handle secrets, write authentication material, and execute downloads under a misleadingly narrow description.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill instructs use of sensitive credentials from .env and browser cookies, then caches a token locally, without clear warning, scoping, storage protections, or lifecycle controls. This increases the risk of credential leakage, token theft, accidental commit of secrets, and unauthorized reuse by other local processes or users.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The script persists an access token in .tga-token on disk in plaintext. If the skill directory is accessible to other local users, backup tools, or accidentally committed/shared, the token could be reused to access TGA data without reauthentication.

Static analysis

No suspicious patterns detected.